The bottom line is that Java is, and has been for some time, a far better platform for Linux development than Mono.
If so, then why do are there no popular desktop apps written in it? At all?
I could say something annoying in a flamebait style like:
FACT: Java on the desktop is dead, it had its chance, several times, and dropped them all.
but..... nah.
The free software community has not produced anything like it. Cloning such technologies has been done before, and now it's happening again. If you really think Mono is evil and Java is our saviour, then I hope you have been hacking on the Java/GTK bindings, writing kickass software using them, and showing the world how it's meant to be done.
these guys clearly have an agenda of creating interest and demand for C# on Linux
Uh, no. The guy pushing Mono is Miguel, Dashboard is written by Nat. Large parts of Ximian are not using Mono, or not sure about it. Nonetheless, Nat decided to use Mono because he wanted to try it out, and decided he liked what he saw.
This is a big win for Microsoft, and pretty much of no help to Linux
I fail to see how having a language that doesn't suck and is good for writing desktop apps in is "no help at all to Linux". We have to move beyond C and C++ sometime, you know.
Mono is always going to be on shaky ground legally
What, just like Linux itself? Need I remind you that Linux is a reimplementation of a once proprietary OS written in a once proprietary language?
If they wrote their application for Java instead.....[snip]
Yes, but they didn't. Java had its chance at being the choice of desktop apps, and failed it. I'm still not sure why Java is any better than C# - both languages/environments have had to have their compilers and class libraries cloned by the free software community, except that Java is not ratified by ISO.
Essentially Ximian are developing this software to grab some market share for something they hope will become big
Er, no. Why don't you read up on how it got started? It started as Nat trying something he thought would be cool, and then a load of people helping out. There are enough conspiracy theories as it is, no need to invent more.
Think before you endorse C#.
The people writing this stuff aren't stupid. I can tell you, they are thinking about C#, and have decided that most peoples fears are groundless. That doesn't mean it's risk free! Oh no. It does mean that they think it's no more risky than any other part of free software is (cloning a UNIX kernel? reverse engineering Windows? GPLd MP3 players?)
Re:emacs: been there done that
on
Nat Demos Dashboard
·
· Score: 2, Interesting
Oh, these guys are perfectly aware of the RA, but they are trying to make something better. In particular, it has better indexing abilities, and far better integration with apps that people actually use.
You don't have a problem with people trying to do the RA better, right?
Agreed, especially true because the full Dashboard system is written in a range of languages - the plugins/backends tend to use whatever languages are best for integration with the software. It's certainly not a pure.NET app, not by any stretch.
Of course, this is all assuming that Gates won't sneak some GPLed code into Windows now, and just claim that we stole it from him instead... Which is a decidedly more frightening prospect.
Haha, as somebody who has seen code that almost certainly resembles the innards of Windows (MIDL output, MFC etc) I can tell you that in no way, shape or form could people ever mix the two up. All you have to do to sniff out Microsofts code is look for moronic types like LPVOID, INT, and parameter names like "lpszStr". Fortunately the horror of Hungarian notation is nowhere to be found on Linux (phew!)
Actually, if you take Apples own published numbers and numbers from IDG on users of MacOS X and Linux, you find that there are approximately 3-4 times as many desktop Linux users than Mac users.
This is not entirely surprising, although it depends on how IDG measured what a "Linux desktop" user is (what if they dual boot etc). Linux generally has a very low barrier to entry, whereas Macs have a very high barrier to entry.
Now, instead of flaming the article and anybody who dares comment on it, why don't you back your position up with some numbers, like the people who are claiming the Linux is about to (or already has) overtaken MacOS X are doing?
What a load of rubbish. It's the default in Red Hat Rawhide because it's the default in the next version of Gnome, that was not a decision red hat made.
As for the reasons why it is the default in Gnome, that might have something to do with the fact that MPG is co-operative and convinced the gnome release team he was aligned with their goals, as opposed to the Galeon team, who did not.
Once you realize that this maintains your filesystem structure but turns your files into Swiss cheese (this happened to me several times) you'll switch to ext3 or XFS.
That note just says what is common knowledge already, that if you want an actual crash proof system you need data journalling. I know very, very few people who use this, because it effectively means writing everything twice, which kills performance. That doesn't reflect upon ReiserFS at all, many filing systems don't support data journalling because it is actually quite a niche feature.
It's also wrong to assume that data journalling == no problems in the event of a crash. That's not the way it works.
Hans is recklessly pursuing optimization of one aspect of filesystem performance. I honestly believe DARPA should reconsider funding his current track of work.
I think you should wait and see how it actually behaves in the real world, before slamming him or his work (as an anonymous cowards, I might add).
You're saying they are? Last time I checked, which was perhaps a few months ago now, they were not scalable and entirely pixmap based. AFAIK no major update has come out since then.
Well, I think GNOME can use SVG graphics pretty much throughout the desktop these days, although SVG icons tend to look bad at low resolutions so not many people use it.
Having said that widget toolkits with containment based layouts like GTK and Qt are much better for this sort of thing. Traditional Win32 widgets/windows have no concept of geometry management, meaning that they are hard to make resizable and don't deal well with text changing their size as can happen with odd font sizes and internationalized text.
I would guess this is what puts Linux at a large advantage over Windows here, rather than any part of the window manager (which has no effect on font size, I would note).
Being based on OpenGL, PDF, and making extensive use of TrueType fonts, I was under the impression that Quartz and MacOS X were aptly suited for this sort of use.
Every OS makes use of TrueType fonts.
IIRC, essentially the entire UI is vector graphics (being done by OpenGL and all), so Apple might have this covered.
You do not recall correctly. The Aqua gui is entirely pixmap based, the widgets aren't even scalable (which has caused the safari team some grief).
Re:To bad Evolution probably wont support it
on
Kroupware Komplete
·
· Score: 1
Not in my experience, it seems to be mostly 40%/40% with the remainder using enlightenment/openbox/whatever WM. Especially since Red Hat 8 came out, there has been much greater usage of gnome around here I've noticed.....
[1] I just noticed (pause for laughter) that the current technologies page is now highlighting Applescript as a top-level component of the OS, on the same conceptual level as Aqua. Imagine that, an OS that considers scripting and automation to be equally as important as the GUI. That's a nice balance.
I think you've never played with OLE Automation, which is easily one of the most sophisticated scripting frameworks in the world. Unfortunately, it's not been well marketed, and many people are not aware of how to tap its full power.
It was never a law, corporate entities cannot invent laws, however region encoding your players was a part of the contract you had to sign in order to get the decryption keys (iirc). Effectively the DVD CCA (?) used the fact that they held all the keys to enforce arbitrary rules.
Since the LiViD/piracy scene located all they keys and broke the decryption system, the contracts effectively became meaningless.
The primary one for me is that it's incompatible with Linux (and other free software). I don't want to use Windows just to play my music, but if I can play my DRM encoded music on Linux then clearly I can circumvent it.
I don't have any real issues with enforcing payment for copyrighted works, although given the general brokenness of the current distribution model (I never did get that Chicane CD my family tried to get for Christmas) it'd make sense to try and come up with a model that's more compatible with it.
Then, you want to snarf emacs (for writing code) and Glade (for designing guis). Python has the kind of easy syntax and clear documentation that lets you write code really fast, and it also has lots of nice features.
A lot of the Red Hat are written using this combination, as well as yum (a bit like apt), Straw (rss aggregator) etc - it's the VB for Linux, except good.
Re:font rednering
on
QT 3.2 Released
·
· Score: 3, Informative
FreeType is cross platform though;)
FreeType however does not do the relevant layout you need for all international scripts. GTK uses Pango for this purpose, TrollTech rolled their own. FreeType is just involved with rendering the glyphs, there is quite a bit more involved to completely display unicode text.
Well, it seems it's not comparing formats, it's comparing encoders for AAC, to find out which is the best. The "QuickTime" encoder (exactly where it comes from, or what it is doesn't seem to be mentioned) won. If you aren't interested in AAC audio though, this set of tests doesn't really interest you.
QuickTime won? Surely some mistake, QuickTime is a container format. It can't possibly win an audio codec listening competition, as it's not a codec. If the/. story is transcribed correctly, then that already trashes the credibility of this story.
Windows offers a powerful infrastructure for distributing DRM encrypted media. Apple provide a similar infrastructure (but seem to keep it mostly to themselves). Linux has no such thing - even if it had 40% of the market, I doubt they would be targetting it.
Slashdot Mirror
If so, then why do are there no popular desktop apps written in it? At all?
I could say something annoying in a flamebait style like:
FACT: Java on the desktop is dead, it had its chance, several times, and dropped them all.
but ..... nah.
The free software community has not produced anything like it. Cloning such technologies has been done before, and now it's happening again. If you really think Mono is evil and Java is our saviour, then I hope you have been hacking on the Java/GTK bindings, writing kickass software using them, and showing the world how it's meant to be done.
When computers choose the software for us, of course ;)
Just kidding. It's more likely to be economics that chooses.
Uh, no. The guy pushing Mono is Miguel, Dashboard is written by Nat. Large parts of Ximian are not using Mono, or not sure about it. Nonetheless, Nat decided to use Mono because he wanted to try it out, and decided he liked what he saw.
This is a big win for Microsoft, and pretty much of no help to Linux
I fail to see how having a language that doesn't suck and is good for writing desktop apps in is "no help at all to Linux". We have to move beyond C and C++ sometime, you know.
Mono is always going to be on shaky ground legally
What, just like Linux itself? Need I remind you that Linux is a reimplementation of a once proprietary OS written in a once proprietary language?
If they wrote their application for Java instead.....[snip]
Yes, but they didn't. Java had its chance at being the choice of desktop apps, and failed it. I'm still not sure why Java is any better than C# - both languages/environments have had to have their compilers and class libraries cloned by the free software community, except that Java is not ratified by ISO.
Essentially Ximian are developing this software to grab some market share for something they hope will become big
Er, no. Why don't you read up on how it got started? It started as Nat trying something he thought would be cool, and then a load of people helping out. There are enough conspiracy theories as it is, no need to invent more.
Think before you endorse C#.
The people writing this stuff aren't stupid. I can tell you, they are thinking about C#, and have decided that most peoples fears are groundless. That doesn't mean it's risk free! Oh no. It does mean that they think it's no more risky than any other part of free software is (cloning a UNIX kernel? reverse engineering Windows? GPLd MP3 players?)
You don't have a problem with people trying to do the RA better, right?
Agreed, especially true because the full Dashboard system is written in a range of languages - the plugins/backends tend to use whatever languages are best for integration with the software. It's certainly not a pure .NET app, not by any stretch.
Haha, as somebody who has seen code that almost certainly resembles the innards of Windows (MIDL output, MFC etc) I can tell you that in no way, shape or form could people ever mix the two up. All you have to do to sniff out Microsofts code is look for moronic types like LPVOID, INT, and parameter names like "lpszStr". Fortunately the horror of Hungarian notation is nowhere to be found on Linux (phew!)
This is not entirely surprising, although it depends on how IDG measured what a "Linux desktop" user is (what if they dual boot etc). Linux generally has a very low barrier to entry, whereas Macs have a very high barrier to entry.
Now, instead of flaming the article and anybody who dares comment on it, why don't you back your position up with some numbers, like the people who are claiming the Linux is about to (or already has) overtaken MacOS X are doing?
As for the reasons why it is the default in Gnome, that might have something to do with the fact that MPG is co-operative and convinced the gnome release team he was aligned with their goals, as opposed to the Galeon team, who did not.
That note just says what is common knowledge already, that if you want an actual crash proof system you need data journalling. I know very, very few people who use this, because it effectively means writing everything twice, which kills performance. That doesn't reflect upon ReiserFS at all, many filing systems don't support data journalling because it is actually quite a niche feature.
It's also wrong to assume that data journalling == no problems in the event of a crash. That's not the way it works.
Hans is recklessly pursuing optimization of one aspect of filesystem performance. I honestly believe DARPA should reconsider funding his current track of work.
I think you should wait and see how it actually behaves in the real world, before slamming him or his work (as an anonymous cowards, I might add).
You're saying they are? Last time I checked, which was perhaps a few months ago now, they were not scalable and entirely pixmap based. AFAIK no major update has come out since then.
Having said that widget toolkits with containment based layouts like GTK and Qt are much better for this sort of thing. Traditional Win32 widgets/windows have no concept of geometry management, meaning that they are hard to make resizable and don't deal well with text changing their size as can happen with odd font sizes and internationalized text.
I would guess this is what puts Linux at a large advantage over Windows here, rather than any part of the window manager (which has no effect on font size, I would note).
Every OS makes use of TrueType fonts.
IIRC, essentially the entire UI is vector graphics (being done by OpenGL and all), so Apple might have this covered.
You do not recall correctly. The Aqua gui is entirely pixmap based, the widgets aren't even scalable (which has caused the safari team some grief).
Not in my experience, it seems to be mostly 40%/40% with the remainder using enlightenment/openbox/whatever WM. Especially since Red Hat 8 came out, there has been much greater usage of gnome around here I've noticed.....
I think you've never played with OLE Automation, which is easily one of the most sophisticated scripting frameworks in the world. Unfortunately, it's not been well marketed, and many people are not aware of how to tap its full power.
Since the LiViD/piracy scene located all they keys and broke the decryption system, the contracts effectively became meaningless.
I don't have any real issues with enforcing payment for copyrighted works, although given the general brokenness of the current distribution model (I never did get that Chicane CD my family tried to get for Christmas) it'd make sense to try and come up with a model that's more compatible with it.
Yes, GObject is over complex. Regardless, it's useful anyway. Find more info here.
And if I remember rightly, didn't he get enough donations to by like 5 iPods or something crazy? Pretty harsh to be trying to get more money now.
Then, you want to snarf emacs (for writing code) and Glade (for designing guis). Python has the kind of easy syntax and clear documentation that lets you write code really fast, and it also has lots of nice features.
A lot of the Red Hat are written using this combination, as well as yum (a bit like apt), Straw (rss aggregator) etc - it's the VB for Linux, except good.
FreeType however does not do the relevant layout you need for all international scripts. GTK uses Pango for this purpose, TrollTech rolled their own. FreeType is just involved with rendering the glyphs, there is quite a bit more involved to completely display unicode text.
Sure, but whos is it? Did Apple write it themselves? Did they buy it? Is it available separately?
Well, it seems it's not comparing formats, it's comparing encoders for AAC, to find out which is the best. The "QuickTime" encoder (exactly where it comes from, or what it is doesn't seem to be mentioned) won. If you aren't interested in AAC audio though, this set of tests doesn't really interest you.
QuickTime won? Surely some mistake, QuickTime is a container format. It can't possibly win an audio codec listening competition, as it's not a codec. If the /. story is transcribed correctly, then that already trashes the credibility of this story.
It includes Windows too - google for "secure audio path"
Windows offers a powerful infrastructure for distributing DRM encrypted media. Apple provide a similar infrastructure (but seem to keep it mostly to themselves). Linux has no such thing - even if it had 40% of the market, I doubt they would be targetting it.